All good points and businesses enforcing this is only a good thing. I do hope that the next step for GDPR will be educating citizens and stopping companies from hiding behind hundreds of pages of privacy policy. Some have done a good job, but it is quite rare.
Honestly I think it's very complex, and particularly for people without an exposure to a commerce or data science environment, it's very challenging to understand how powerful data capture (particularly in the context of an online environment) is.
Particularly in the age of TV and print advertising, the misguided view of "marketing doesn't work on me" was prevalent, but particularly since the marketing paradigm shift I think that view has grown and is, ironically, even more misguided now given the power that tracking enables.
When retailers can build incredibly detailed customer profiles touching on all demographic elements (age, location, current location, estimated salary, political affiliation, contact information, opt in status and preferences, gender), buying behaviour (preferred brands, preferred time to purchase, preferred location and store to purchase from, cross-shopping behaviour, brands likely to convert on, preferred products and product types), sentiment analysis (how they feel about the brand or individual stores, what brands they like and dislike, what products they like and dislike and the reasons, what marketing campaigns they prefer, what sort of copy is best to make people transact or have an emotional response), and behavioural analysis (how often they engage with emails/online ads, when is the best time to send campaigns or recommendations to result in engagement and conversion, tracking how long people on average spend on site and how many items they click before deciding to make a purchase, what sort of customer flow is typical, and what they have an affinity for and are likely to convert next on or could be cross-sold on) there is incredibly powerful and highly subtle techniques that can be applied to optimise sales.
When data from social media channels can be cross-referenced (e.g. integrating social ads from Facebook or Google by matching data or by developing lookalike audiences) to deliver highly personalised advertisements at the time when people are most susceptible to engage, and techniques can be refined by not only AB-testing (as traditionally done) but personalising ads down to an individual level and optimising all aspects of the content for an individual (time you see it, what you see, what the copy is, what the model is like, which product it is and what colours are in the ad) without people even being aware that it's happening, it's very hard in my opinion for people to be properly educated on just how powerful and pervasive data capture can be.
Furthermore, it's very difficult for people to really get a sense of how their own behaviour is influenced by these techniques because of how distant cause and effect can be; a man browsing a bag for a certain brand on Facebook may not realise that this played a direct and subconscious role in purchasing a different handbag for a competing brand on a different website after seeing an image of the handbag held by a male model. The effect is so distanced from what can be measurably shown to have played a significant role in triggering it that from a consumer perspective it's incredibly difficult to understand how the role it actually played (and could be easily dismissed as 'being my own conscious decision' instead of powerful marketing and data-techniques being applied to drive you to that decision).